Biotic and microsite factors affecting Pinus albicaulis establishment and survival
by Ward Wells McCaughey
A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in
Biological Sciences
Montana State University
© Copyright by Ward Wells McCaughey (1990)
Abstract:
To culture a major high elevation tree, we need information on biotic and microsite factors affecting
whitebark pine (Pinus albicaulis) seed survival, emergence, and seedling establishment. This thesis
summarizes the results of the first 2 years of a long-term field study designed to evaluate physical and
biological factors affecting whitebark pine establishment and survival. Predator effects on seed survival
were estimated by recording seedling emergence under four levels of predator exclusion (exclude birds
and rodents, exclude rodents only, exclude birds only, and exclude no predators). Microsite effects
were evaluated by recording seedling emergence on mineral, litter, and burned seedbeds, under shade
cover (0%, 25%, and 50%), and for buried (5 cm) and surface-sown seeds.
Rodents ate or removed all the surface-sown seeds and most of the buried seed they had access to.
Birds took neither surface-sown nor buried whitebark pine seeds.
Whitebark pine seeds with delayed germination, those that laid dormant over two winters, had higher
emergence rates than seeds that germinated after only one winter stratification period. Emergence rates
of buried seeds was significantly greater than for surface-sown seeds. Emergence of surface-sown
seeds preceded buried seeds. First-year emergence on mineral soil was higher than on litter or burned
seedbeds; however, there was no difference in numbers of second-year emergents among seedbed
conditions. Shading improved emergence of both first- and second-year seeds.
Insolation, drought, and rodents were the primary agents affecting survival of whitebark pine seedlings.
Insolation mortality occurred in late June and early July of both measurement years. It was followed by
drought mortality which ended in late August. Shade cover decreased insolation mortality and
increased drought mortality. Drought mortality was higher than insolation mortality. Seedling losses
due to animal damage were minimal and sporadic. More seedling mortality occurred during the winter
than during the second growing season.
If whitebark seeds are planted, seedling emergence including first-year and delayed emergents, may be
highest on shaded mineral seedbeds. BIOTIC AND MICROSITE FACTORS AFFECTING FINDS
ALBICAULIS ESTABLISHMENT AND SURVIVAL
by
Ward Wells McCaughey
!
A thesis submitted in partial fulfillment
of the requirements for the degree
of
Doctor of Philosophy
in
Biological Sciences
MONTANA STATE UNIVERSITY
Bozeman, Montana
June, 1990
ii
APPROVAL
of a thesis submitted by
Ward Wells McCaughey
This thesis has been read by each member of the
thesis committee and has been found to be satisfactory
regarding content, English usage, format, citations,
bibliographic style, and consistency, and is ready for
submission to the College of Graduate Studies. .
ft-INuxItV
Date
”I
Chairperson, Graduate Committee
Approved for the Major Department
17 Moy IcI0IO____
Date
Head, Major Department
Approved for the College of Graduate Studies
Date
Graduate* Dean
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of
the requirements for a doctoral degree at Montana State
University, I agree that the Library shall make it
available to borrowers under rules of the Library.
I
further agree that copying of this thesis is allowable
only for scholarly purposes, consistent with "fair use" as
prescribed in the U.S. Copyright Law.
Requests for
extensive copying or reproduction of this thesis should be
referred to University Microfilms International, 300 North
Zeeb Road, Ann Arbor, Michigan 48106, to whom I have
granted "the exclusive right to reproduce and distribute
copies of the dissertation in and from microfilm and the
right to reproduce and distribute by abstract in any
format."
iv
ACKNOWLEDGMENTS
I thank the U.S. Forest Service, Intermountain
Research Station for the financial support of this project„
I especially thank Wyman Schmidt, Project Leader, for his
continued encouragement, support, and advice in all phases
of my Ph.D. program.
I am thankful to Tad Weaver, my
major professor, for his help during the initial planning
stages and for his guidance throughout the project.
I thank Dick Kracht for helping me locate the
experimental site and the Gallatin National Forest for the
use of Forest Service land and facilities.
I also thank others who helped with various stages of
my program.
Wally Gladstone, Art Stokes, Mark Bachelor,
and Vivek Karnik who helped with the field establishment
phase and initial data input.
Kathy McDonald's formatting
and typing skills are greatly appreciated.
My academic committee— Jarvis Brown, Jim Pickett, and
Dave Cameron— provided support and advice on my studies
and research.
I thank my wife, Cathy, and our children, Alex,
Mandy, and Travis, for their patience, support, and
encouragement throughout this long project.
V
TABLE OF CONTENTS
Page
LIST OF TABLES .......................... ......... . ^
vii
LIST OF FIGURES........ ..... .............. ..______ __o
ix
ABSTRACT.... ........................
xi
I.
INTRODUCTION.................... ..... ..........
1
2.
METHODS ....................................
6
3.
Study Area..............
Study Design. ...........
Treatment Descriptions...........................
Predator Exclusion . ........... .............. .
Shade.............. ...................... ..
Seedbed............................. .........]
Sowing Depth................. .................
Plot and Seed Layout........................... []]
Measurements.....................
*’*’*’
Data Analysis........ ........-..... ..............
21
H
-14
16
RESULTS AND DISCUSSION... ..........
iq
•
•
•
•
e
e
e
e
e
e
o
e
e
e
e
e
e
e
e
e
e
o
e
o
o
e
e
e
e
Seed Predation......
Surface-Sown Seed........
Buried Seed........................
Emergence........................................ ]
Years.........
Predator Exclusion..............
Shade Cover.........
^
Seedbed Condition........
Sowing Depth.....
Replicate................................
Interactions.............
1988 Interactions.....................
1989 Interactions..... .....................
Delayed Emergence......................
Predator Exclusion......
Shade Cover.......
Sowing Depth. ........................
Replicate.............................
Interactions..................................
6
g
g
g
10
19
19
20
24
24
27
30
31
33
34
35
36
39
41
44
44
45
45
45
vi
TABLE OF CONTENTS, (continued)
4.
Mortality...... ............................. ....
Subsurface Soil Temperatures.........
48
57
CONCLUSIONS.............. ... ...................
61
Seed L o s s e s . ..... .
Emergence ....................
Mortality........................ ::::::::::::::::
^9
^
LITERATURE CITED.................
68
APPENDIX.......
74
*
®
*
®
e
e
e
e
e
e
e
e
e
®
e
e
*
e
«
e
e
e
e
e
e
e
U
JL
vii
LIST OF TABLES
Table
Page
1.
Factors and Factor Levels.........
9
2.
Species List of Rodents Trapped
on Whitebark Pine Study Area in
1987 and 1988.... ...............
21
Emergence of Whitebark Pine for
Exclude Birds Only and Exclude
None Treatments in 1988 and 1989.
23
Effect of Year and Biotic and
Microsite Factors on Emergence of
Whitebark Pine: ' An ANOVA.......
25
Effect of Biotic and Microsite
Factors and Their Interactions
on 1988 Emergence of Whitebark
Pine: An ANOVA..................
26
Effect of Biotic and Microsite
Factors and Their Interactions
on 1989 Emergence of Whitebark
Pine : An ANOVA...................
26
Percent Emergence of Whitebark Pine
as Affected by Predator Exclusion,
Shade Cover, Seedbed Condition,
Sowing Depth, and Replicate on
Treatments Excluding Rodents and
Birds (EA) and Excluding Rodents
Only (ER) for 1988 and 1989 First-Year
and 1989 Delayed Emergents From the
1987 Sowing................. ...............
29
Effect of Biotic and Microsite
Factors and Significant Interactions
on Delayed Emergence of Whitebark
Pine: An ANOVA.....................
43
3.
4.
5.
6.
,7.
8.
viii
LIST OF TABLES
(continued)
Table
9.
10.
11.
12.
13.
14.
15.
Page
Mortality of 1988 and 1989 Whitebark
Pine Emergents on Mineral and Litter
Seedbeds Under 0%, 25%, and 50% Shade
52
Maximum Surface Temperatures Recorded
on Mineral, Litter, and Burned Seedbeds
Under 0%, 25%, and 50% Shade Cover
in 1988 and 1989......... .
53
Minimum-Maximum Soil Temperatures at a
Depth of 2.5 cm on Mineral and Litter
Seedbeds Under 0%, 25%, and 50%
Shade Cover— 1988 and 1989 .........
58
Precipitation based on inches of
accumulated water per month for the
hydrologic years 1987, 1988, and 1989
at the Canyon weather station in
Yellowstone National Park, Wyoming.... .
75
Precipitation based on inches of
accumulated water per month for the
hydrologic years 1987, 1988, and 1989
at the Mill Creek weather station on
the Gallatin National Forest...... .
76
1988 analysis of variance results
showing significance of biotic and
microsite factors and two-factor
interactions on cumulative percent
emergence (arc sine of the square
root of proportion transformation)
of whitebark pine for each recording date...
77
1989 analysis of variance results
showing significance of biotic and
microsite factors and two-factor
interactions on cumulative percent
emergence (arc sine of the square
root of proportion transformation)
of whitebark pine for each recording date...
78
ix
LIST OF FIGURES
Figure
1.
2.
3.
4.
5.
6.
7.
8.
Page
Study Site Location. Gallatin
National Forest. Section 14,
Township 9 S , Range 9 E,
Montana Principal Meridian..................
.7
Schematic Layout of Study Design
Showing All the Treatment
Combinations in One of Three
Replications ............. ..............
12
Schematic Layout of 1987 and 1988
Surface and Buried Seed Locations......... .
13
Cumulative Number of Whitebark Pine
Emergents From Buried and Surface
Sown Seeds in 2 Years.......................
28
Soil Water in Top 5 cm
of Soil on Mineral and Litter
Seedbeds Under 0%, 25%, and
50% Shade Cover in 2 Years...... ..........
32
Percent Moisture in Top 5 cm of Mineral
Soil by Replicate for 1988 and 1989........
35
Two-Way Interactions of the 1988
Biotic and Microsite Factors
Affecting Whitebark Pine
Emergence: Predator Exclusion by
Shade Level (A), Seedbed Condition
(B), Sowing Depth (C), and Percent
Shade Cover by Seedbed (D) ..................
37
Two-Way Interactions of the 1989
Biotic and Microsite Factors
Affecting Whitebark
Emergence: Predator Exclusion by
Shade Level (A), Seedbed Condition
(B), and Seedbed Condition by Sowing
Depth (C)
40
X
LIST OF FIGURES
(continued)
Figure
9.
10.
11.
12.
13.
14.
■ 15.
Page
Two-Way Interactions of the 1989
Biotic and Microsite Factors
Affecting Delayed Emergence
of Whitebark Pine: Predator Exclusion
by Shade Level (A), and Sowing Depth
(B) and Seedbed Condition by Sowing
Depth (C)................. ..........
46
First-Year Mortality Counts of Whitebark
Pine Seedlings by Cause Over Time 1988 (A)
and 1989 (B).....................
49
Survival of Whitebark Pine Seedlings
Germinated From Buried
and Surface SownSeeds.....................
51
Percent Moisture in Top 5 cm of
Soil on Mineral and Litter Seedbeds
Under 0%, 25%, and 50% Shade
in 1988.................................
55
Percent Moisture in Top 5 cm of
Soil on Mineral and Litter Seedbeds
Under 0%, 25%, and 50% Shade
in 1989.......................
56
Mean Minimum-Maximum Soil Temperatures
at a Depth of 2.5 cm on Mineral and
Litter Seedbeds Under 0%, 25%, and
50% Shade Cover by Date for 1988...........
59
Mean Minimum-Maximum Soil Temperatures
at a Depth of 2.5 cm on Mineral and
Litter Seedbeds Under 0%, 25%, and
50% Shade Cover by Date for 1989..... .....
60
xi
ABSTRACT
To culture a major high elevation tree, we need
information on biotic and microsite factors affecting
whitebark pine '(Pinus albicaulis) seed survival, emergence,
and seedling establishment.
s
This thesis summarizes the results of the first
2 years of a' long-term field study designed to evaluate
physical and biological factors affecting whitebark pine
establishment and survival. Predator effects on seed
survival were estimated by recording seedling emergence
under four levels of predator exclusion (exclude birds and
rodents, exclude rodents only, exclude birds only, and
exclude no predators). Microsite effects were evaluated
by recording seedling emergence on mineral, litter, and
burned seedbeds, under shade cover (0%, 25%, and 50%), and
for buried (5 cm) and surface-sown seeds.
Rodents ate or removed all the surface-sown seeds and
most of the buried seed they had access to. Birds took
neither surface-sown nor buried whitebark pine seeds.
Whitebark pine seeds with delayed germination, those
that laid dormant over two winters, had higher emergence
rates than seeds that germinated after only one winter
stratification period. Emergence rates of buried seeds
was significantly greater than for surface-sown seeds.
Emergence of surface-sown seeds preceded buried seeds.
First-year emergence on mineral soil was higher than on
litter or burned seedbeds; however, there was no difference
in numbers of second-year emergents among seedbed
conditions. Shading improved emergence of both first- and
second-year seeds.
Insolation, drought, and rodents were the primary
agents affecting survival of whitebark pine seedlings.
Insolation mortality occurred in late June and early July
of both measurement years. It was followed by drought
mortality which ended in late August. Shade cover
decreased insolation mortality and increased drought
mortality. Drought mortality was higher than insolation
mortality. Seedling losses due to animal damage were
minimal and sporadic. More seedling mortality occurred
during the winter than during the second growing season.
If whitebark seeds are planted, seedling emergence
including first-year and delayed emergents, may be highest
on shaded mineral seedbeds.
I
CHAPTER I
INTRODUCTION
Whitebark pine (Finns albicaulis Engelm.) is a high
elevation species which ranges from northern British
Columbia to south-central California and from the Pacific
coast range to the Wind River range in Wyoming (Critchfield
and Little 1966).
Whitebark pine communities comprise 10%
to 15% of the forested landscape (1.2 million hectares) in
the Rockies from western Wyoming to the Canadian border
(Arno 1986).
It forms only a minor component of forest
communities which are commercially harvested, but is found
in pure stands immediately below timberline.
Pure stands
may be either woodlands with widely spaced diffuse crowned
trees or "krummholz" with a low flagged form (Arno and
Weaver 1990).
Whitebark pine is found, in a variety of habitats and
grows with several tree species including subalpine fir
(Abies lasiocarpa), Engelmann spruce (Picea engelmannii),
and lodgepole pine (Pinrus contorta) in the Yellowstone
ecosystem (McCaughey and Schmidt 1990; Pfister et al.
1977; Weaver and Dale 1974).
Understory associates are
listed by Forcella (1977), Pfister et al. (1977), and
Weaver and Dale (1974).
Vaccinium scoparium is the most
LI
2
abundant understory species in pure or nearly pure stands
of whitebark pine (McCaughey and Schmidt 1990; Weaver and
Dale 1974) .
Throughout its range whitebark pine is important for
esthetics, watershed protection, wildlife food and cover
(Eggers 1986; Kendall 1983), and ornamental planting.
Whitebark pine stands provide cover for a variety of
plants and animals in timberline and subtimberline zones
(Arno and Hoff 1989).
Its seeds are an important food
source for grizzly (Ursus arctos horribilis) and black
bear (Ursus americanus) (Craighead et al. 1982; Kendall
1983; Knight et al. 1987) and a supplemental food source
for birds (Tomback 1982; VanderWall and Hutchins 1983) and
other small animals (Hutchins and banner 1982; McCaughey
and Schmidt 1990) .
Whitebark pine is threatened by mountain pine beetle
(Dendroctonus ponderosae), white pine blister rust
(Cronartium ribicola), and fire suppression (Amman 1982;
Arno 1986).
Mountain pine beetle and white pine blister
rust are direct killers of whitebark pine.
Fire
suppression reduces available habitat by allowing its
replacement more competition-tolerant subalpine fir and
Engelmann spruce.
M
3
Management for .the survival of whitebark pine forests
will require management of plant competitors, insects,
disease, and the tree itself.
For example:
I. Wildfires
or prescribed burns may be needed to maintain whitebark in
areas where it is serai.
2. Stand conditions in lower
lodgepole forests may be manipulated to reduce the effect
of beetles on whitebark pine.
This is because mountain
pine beetle populations build to epidemic proportions in
lodgepole forests and sweep up into high elevation
whitebark stands (Arno and Hoff 1989) where they are
unable to sustain themselves due to severe climatic
conditions.
3. Exotic diseases, especially white pine
blister rust, must be managed.
4. Where competition or
disease mortality cannot be reduced, regeneration must be
increased to compensate for losses.
Studies of the regeneration process of whitebark pine
will contribute to the long-term survival of the species
by improving regeneration.
Little is known about seed
production (Kendall 1983; Weaver and Forcella 1986) or
regeneration processes of whitebark pine under natural or
artificial conditions (Eggers 1985).
Weaver and Dale
(1974) recorded whitebark pine regeneration rates in
undisturbed climax communities of whitebark pine.
While
seedlings germinated in meadows, openings created by
disturbances, and closed forests, those not occurring in
openings (large or small) rarely produce cones (Weaver et
Jl
4
al. 1990).
The germination percent of seeds cached by the
nutcracker is unknown but of seedlings germinating in
nutcracker caches 56% survived the first year but only 25%
survived the third year (Tomback 1982).
Published
information is sparse on later growth autecology of
whitebark pine, including mechanisms of flowering and
fruiting, cone production (Weaver and Forcella 1986), seed
characteristics and dissemination (banner 1982; Tomback
x
1982), vegetative reproduction, growth and morphology,
rooting (Jacobs and Weaver 1990) , shade tolerance (Arno
and Hoff 1989), longevity, and phenology.
Whitebark pine
has not been considered a timber production species
because of its slow growth and generally poor form (Arno
and Hoff 1989; Weaver et al. 1990).
This study was designed to determine the effects of
biotic and microsite factors on seed survival, germination
emergence, and first year survival.
The design includes
treatments that are directly applicable to silvicultural
practices.
Five specific objectives of this study were
to:
I.
Determine differences in seed loss due to bird
and small mammal predators when seed are surface sown
(simulating unusual but conceivable tree dispersal) and
when seed are buried 2 to 4 cm in soil (simulating burial
by Clark's nutcracker [Nucifraga Columbiana]).
I r
5
2„
Compare seedling emergence and establishment from
surface-sown seeds and seeds buried 2 to 4 cm in soil.
3.
Compare seedling emergence and establishment on
mineral, litter, and burned seedbeds.
4.
Compare emergence and establishment under 0%,
25%, and 50% shade cover.
5.
Record seedling survival rates across seedbeds
and shade treatments.
Compare between-year differences in emergence
across seedbed and shade treatments.
il
6
CHAPTER 2
METHODS
Study Area.
The experimental site was identified as an Abies
lasiocarpa - Pinus albicaulis/Vaccinium scoparium habitat
type (Pfister et al. 1977) occupied mainly by lodgepole
pine.
Most of the area appears to be on an inceptisol/
This soil has a 6 to 8 cm thick cambic "B" horizon between
the A and
and
horizons.
The soils on a small
portion of the study area are were identified as Typic
Cryorthent, sandy skeletal being well drained (Soil Survey
Staff 1975).
This soil has a 12 cm thick "A" horizon
overlying C^ and C^ horizons.
5 cm thick "0" horizon.
5.5.
Both soil types have a 3 to
Soil pH values range from 4.7 to
The elevation is 2,652 m MSL with 0% to 25% slopes
and a northeast aspect.
The study area is located in section 14, township 9
south, range 9 east on the Gardiner Ranger District of the
Gallatin National Forest.
It is north of Yellowstone
National Park (Figure I), and near the southwestern corner
of the Absaroka Beartooth Wilderness approximately 8.8 air
kilometers east of Gardiner, MT.
7
MONTANA
Gardln
S t u d y Site
t Yellowstone
- — I National
Park
IDAHO
WYOMING
Figure I.
Study site location. Gallatin National Forest.
Section 14, township 9 S , range 9 E , Montana
Principal Meridian.
Study plots were established on a 6 hectare clearcut
which is connected on the east to a large clearcut
(20 hectare) called the Palmer Coop timber sale.
The
entire area was harvested during the winter of 1985-1986.
Approximately 305 to 358 m^/ha of timber were harvested
with 22 to 33 metric tons/ha of slash left on the site.
The species and volumes harvested were:
live lodgepole
pine - 75%, dead lodgepole - 13%, Engelmann spruce - 4%,
subalpine fir - 4%, and whitebark pine - 4%.
The study
area is bordered by a mature forest of similar composition
on the south and west, and a forest with 20% whitebark
pine to the north.
8
Study Design
A factorial experiment (Table I) was used to
determine the effects of seed predators, light levels,
seedbed conditions, and seed sowing depths on the
germination and early survival of whitebark pine.
Three
subsites (replicates) were subjectively chosen within the
6 ha clearcut as representative, similar, and suitable for
plot establishment.
The subsites had minimal amounts of
logging slash, large areas of undisturbed litter, and
reasonably represented the overall stand conditions.
Figure 2 is a schematic diagram of one replication of each
predator exclusion - shade cover - seedbed condition sowing depth combination.
Plots were randomly located in
each replicate.
Treatment Descriptions
Predator Exclusion
Four treatments were used to evaluate predation
effects on whitebark pine seed:
exclude birds and rodents,
exclude rodents only, exclude birds only, and exclude
none.
Wire screen was used to exclude seed predators from
the plots while plots exposed to all predators were
un-screened.
Plots protecting seeds from all predators
were completely covered using hardware cloth with 0.63 cm
9
Table Ii
Ti
Factors and factor levels.
Factor
Predator exclusion
a. Exclude birds and rodents (EA)
b. Exclude rodents only (ER)
c . Exclude birds only (EB)
d. Exclude none (EN)
~
Levels
4
2.
Shade level
a . No shade
b. 25 percent shade
c . 50 percent shade
3
3.
Seedbed condition
a. Mineral (1988 analysis)
b. Litter (1988 analysis).
c . Burned (1989 analysis)1
3
4.
Sowing depth
a. Surface-sown
b. Seed buried ( 2 - 4 cm)
2
5.
Replication
3
I
First year results did not include a burned seedbed
treatment.
square holes.
Plots for protecting seed from birds only
were covered by screen with 5 by 7.6 cm wide holes.
Plots
excluding rodents only were enclosed by a 76 cm high fence
of 0.63 cm square mesh hardware cloth.
The rodent fence
was designed to exclude rodents but allow access to avian
predators.
The top of the fence had a 20 cm lip, bent
outward from the plot.
A 15 cm piece of tin flashing was
attached to the underside of the lip to effectively
exclude rodents.
The bottom of screens were buried 10 to
15 cm deep on plots excluding both birds and rodents and
plots excluding rodents only.
The bottom edge of the
buried screen had a 5 cm lip bent outward from the plot to
minimize the chance of rodents tunneling under the screen.
Screening techniques for the control of seed predation
were suggested by Curt Halverson, U.S. Fish and Wildlife
Service, Fort Collins, CO.
Shade
Three shade treatments were used; no shade, 25%, and
50% shade cover.
slatted roofs.
Shade treatments were imposed with
Four 1.8 m tall steel posts were installed
at the corners of an imaginary 1.2 by 2.4 m rectangle
overtopping but slightly to the south of each plot to be
shaded.
A 1.2 by 2.4 m long wood frame was constructed
with 5 by 10 cm lumber and attached to the steel posts
100 cm above the ground.
A 1.2 x 2.4 m section of wood
snow fence was suspended on the wood frame.
The 50% and
25% shade levels were simulated by either leaving all the
wood slats in the snow fence or by removing alternate
slats respectively.
Seedbed
Mineral, litter and burned seedbed treatments were
used to examine emergence and survival of whitebark pine.
Mineral seedbed treatments were located on scarified skid
trails or hand scalped (top 2 to 5 cm of soil) when
I
11
logging scarified areas did not occur on a mineral
treatment location.
Litter treatments consisted of areas
undisturbed by logging.
Prescribed broadcast burns created the burn seedbed
conditions.
2
Because burning was done in confined (15 m )
areas of the clearcut, burned treatment plots were randomly
located within a burned area adjacent to the mineral and
litter seedbed plots for that replicate.
The burn
treatment areas were burned twice in 1987 because high
moisture content of litter and other fine fuels created
poor burning conditions.
The completed treatment resembled
scattered spots of a light surface fire even after two
burns.
Because burning was not completed until late fall
of 1987, no plots were seeded until the fall of 1988.
Sowing Depth
Two sowing depth treatments were used; surface-sown
and buried.
Surface-sown seed were placed on the ground
surface and buried seed were buried 2-4 cm below the
surface level and covered.by the appropriate seedbed
material (i.e ., covered by mineral soil, litter, or ash).
Plot and Seed Layout
Within each subsite, 36 plots were established to
represent all combinations of the four predator exclusion,
three shade, three seedbed, and two sowing depth treatments
(Figure 2).
Plots were rectangular (0.5 x 2.0 m) and
12
SHADE
COVER
Mineral
Litter
Burned
m
qei
I Replication
Figure 2.
Schematic layout of study design showing all
the treatment combinations in one of three
replications. In the field, treatment
combinations were located randomly. Predator
treatments were exclude birds and rodents (EA),
exclude rodents only (ER), exclude birds only.
(EB), and exclude none (EN).
oriented the long direction north-south.
The south half
(0.5 x 1.0 m) of each plot was seeded in the fall of 1987
and the north half was seeded in the fall of 1988.
In 1987, the south half of each plot was subdivided
into 40 subplots measuring 10 by 11 cm (Figure 3).
Within
each subplot two seeds were planted, one surface-sown and
one buried.
The surface-sown seed were placed in the
north half of each subplot.
The buried seed were placed
in the south half of each subplot.
In 1988, the north half of each plot was subdivided
into two halves, a surface-sown half (north) and a buried
half (south), each measuring 49 by 100 cm (Figure 3).
seed sowing design was modified to reduce measurement
The
13
I
1987
Buried Seed
Figure 3.
1988
I
* Surface- Sown Seed
Schematic layout of 1987 and 1988 surface and
buried seed locations.
errors since surface-sown seeds were occasionally moved
short distances due to natural factors such as wind, rain,
and snow and this sometimes made identification of sowing
depth type difficult.
The surface and buried halves were
further subdivided into 48 subplots with each subplot
measuring 6 x 8 cm.
On the surface half, 50 seeds were
sown, one placed on the ground in each of 46 subplots and
two seeds in 2 subplots.
On the buried half, 50 seeds
were placed 2 to 4 cm below the surface level in
46 subplots and two seeds in 2 subplots.
Again, the
buried seed were covered by the appropriate seedbed
material (mineral soil, litter, or burned litter).
1
---------
1
1
I' '
'
I
_
I. I I
'
Il
Il
14
Exactly 5,760 whitebark pine seeds were planted in
1987 and 10,800 in 1988.
The addition of the burned
seedbed condition and an increase from 40 to 50 seeds
accounted for the increased seed numbers in 1988.
All
seeds were x-rayed and only filled seeds were planted.
X-rays do not reduce the germinative capacity or initial
seedling growth of conifer seeds, however, it is unknown
if there are long-term growth effects (Borzan 1973).
Seeds planted in 1987 were collected (seedlot I) in 1985
and stored in sealed plastic bags (relative humidity = 6%
to 8% inside bag) for two years under standard external
conditions (temperatures = -17 to -20°C; humidity = 30%)
until planting.
Seeds were collected (seedlot 2) in the
fall of 1987 and stored under the same conditions as
seedlot I until planting in 1988.
Measurements
Whitebark pine seedlings were counted periodically
throughout spring and summer on all plots.
on June 16 in 1988 and June 9 in 1989.
Counting began
I could not
determine exactly when germination occurred since half the
whitebark seeds were buried and continuous monitoring of
surface-sown seed was prohibitive.
Emergence and emergents
are used to describe germination and to quantify .resultant
seedlings, respectively.
Emergents were counted and
numbers recorded weekly until the first of August and
W
I
15
bimonthly from August to the first of October.
Emergents
were marked with colored plastic toothpicks of different
colors to record the emergence week.
The week of mortality
and its likely cause was recorded for all dead seedlings.
Soil moisture was measured gravimetrically in 1988
and 1989 on 6 of the 24 germination plots at each of the
three replicates.
These six plots comprised one plot from
each combination of mineral and litter seedbed and 0%,
25%, and 50% shade cover.
Soil from the upper 5 cm of the
A horizon was collected in soil cans and sealed for
transport from the field to the laboratory.
Percent soil
moisture was determined by comparing wet and dry weights
(Soil Survey Staff 1975) .
Soil moisture was never measured
on burned seedbeds due to the limited burned treatment
area available.
Subsurface soil temperatures were measured in 1988
and 1989 with Taylor minimum-maximum thermometers at the
same seedbed-shade plots where soil moisture collections
were taken on replicates I and 3.
Soil temperatures in
replicate 2 were measured with temperature probes connected
to electronic microprocessors designed for continuous
collection of environmental conditions.
Problems with
temperature probes caused sporadic and sometimes unreliable
data.
Minimum and maximum soil temperatures were measured
16
at a soil depth of 2.5 cm (the level where seeds were
buried).
Temperatures were measured and recorded weekly
throughout the 1988 and 1989 summers.
Maximum surface temperatures were measured weekly in
1988 and 1989 with wax (Big.Three Industries-tempiI)
pellets which melt at specific temperatures.
Tempils used
for this study were designed to melt at 37.7, 41.1, 45.0,
51.7, 58.9, 65.6, 72;8, 79.4, 86.7, and 93.3 degrees
Celsius.
Tempils were placed on one of the mineral and
litter seedbeds on each of the 0%, 25%, and 50% shade
plots for a total of six plots on each replicate.
Data Analysis
The whitebark emergents/seed planted ratio on each
subplot was used as the dependent variable for analysis of
emergence differences between years, predator exclusion
levels, shade levels, seedbed conditions, sowing depth,
and factor interactions.
Proportion of emergence is
defined as the number of emergents divided by the number
of seeds sown (40 in 1988, 50 in 1989).
Empty plots were
counted as l/4n to .prevent distortion of the analysis by
small numbers (Mosteller and Youtz 1961) . A
transformation, arc sine of the square root of the
proportion of germination, was used to stabilize variation
due to proportions (Snedecor and Cochran 1980) .
17
The statistical analysis system (SAS 1987) was used
to analyze whitebark pine emergence and seedling survival
data.
Analysis of variance was used to test.for
statistical significance of main factors and interactions
on seedling emergence and survival. ANOVA was also used
for evaluation of soil moisture and temperature data.
The
"F" statistic was used to determine the significance of
factors and their interactions on emergence of whitebark
pine.
Multiple comparison procedures were used to analyze
differences between factor levels.
I used the Ryan-Einot-
Gabriel-Welsch multiple F test for equal cell sizes and
the Tukey-Kramer method for unequal cell sizes (SAS 1987) .
All significance tests were done at the "p<0.1" level.
I
chose this high p-value level for two reasons:
(I) regeneration data typically has a high degree of
variation and (2) p-values between 0.05 and 0.1 indicate a
'strong relationship that might otherwise be overlooked.
Predation on whitebark pine seed was analyzed
separately from microsite factors affecting whitebark
germination.
Seed predation was assessed using results
from all four exclusion treatments (BA, ER, EB, and EN).
The EA and ER treatments were used in analysis of variance
to assess whitebark emergence and survival differences
among shade levels, seedbed conditions, and sowing depths.
18
The fact that no seed were lost to predators in these
treatments fully eliminated predation effects and allowed
direct comparisons to be made.
19
CHAPTER 3
RESULTS AND DISCUSSION
Seed Predation
Birds and rodents were the principal potential
predators on whitebark seed considered in this study.
The
Clark's nutcracker, the major bird species consuming
whitebark seed, harvests directly from cones (Hutchins and
banner 1982).
In other studies chipmunks (Eutamia spp.),
deer mice (Peromyscus maniculatus), and golden-mantled
ground squirrels (Spermophilus lateralis) were the
principal rodent consumers (Hutchins and banner 1982;
banner 1980; Tomback 1981).
I considered insects a minor
predator and saw none feeding on or removing whitebark
pine seed.
Surface-Sown Seed
In 1987 and 1988, animals removed 100% of surface-sown
seeds on exclude birds only (EB) and exclude none (EN)
treatments within 5 days after sowing.
No seeds were
removed from exclude birds and rodents (EA) and exclude
rodents only (ER) treatments, indicating that birds were
not randomly searching for whitebark pine seeds and these
screening methods effectively excluded rodents.
Hutchins
20
(1989) believes that random foraging by Clark's nutcrackers
is highly unlikely since their foraging efforts appear to
be directed toward finding their own seed caches.
The
exclosures and shade cover may have discouraged seed
foraging by birds; however, birds, including the Clark's
nutcracker, were observed sitting on exclosures of both ER
treatments and shade structures.
No birds were seen
foraging for seeds on or in the vicinity of any plots.
Birds were observed caching seeds on the study site in
1987 and 1989.
It is assumed that surface-sown seeds on
EB and EN treatments were eaten or removed by rodents
while bird predation was, at most, minimal.
In 1988 and 1989 eight rodent species were trapped on
the study area.
Deer mice represented 54% and southern
red-backed voles 23% of all species caught (Table 2).
Squirrels (Tamiasciurus hudsonicus) clipped whitebark
cones in the adjacent forest but I saw none foraging for
seeds on the study area.
Because of their high frequency
of occurrence and known use of whitebark seeds, deer mice
are probably the main consumers of whitebark seed on this
area.
Buried Seed
Animal predation on whitebark pine seeds buried in
exclude birds only (EB) and exclude none (EN) plots was
shown by depressions on mineral soil and litter seedbeds,
in 1988 and on all seedbed treatments in 1989.
There was
21
Table 2.
Species list of rodents trapped on whitebark
pine study area in 1987 and 1988. Percent
represents the proportion of total sample size
(n=47).
Species _____________ ________ _____________________ Percent
Deer mouse (Peromyscus maniculatus
54
Southern red-backed vole (Clethrionomys gapperi)
23
Masked shrew (Sorex cinereus)
4•
Montane shrew (Sorex monticolus)
2
Montane vole (Microtus montanus)
2
Long-tailed vole (Microtus longicaudus)
9
Heather vole (Phenacomys intermedius)
2
Yellow-pine chipmunk (tamias amoenus
4
no evidence of disturbance at buried seed locations on
exclude rodents only treatments; therefore, it is assumed
that there was no seed predation of buried seeds by birds.
Seeds were untouched on exclude birds and rodents
treatments, indicating that rodent and bird predation was
eliminated by screening.
Pocket gopher (Thomomys
tolpoides) activity was noted in all predation treatments
in 1989; surface-sown seeds were not disturbed (except by
burial with soil brought to the surface) and I have no
evidence of disturbance of buried seed.
Rodents foraged for but did not find all available
buried seeds.
Emergence from buried seeds occurred on
exclude birds only treatments with seeds accessible to
rodents even though all the planting sites were disturbed.
Loss of buried whitebark seeds may have been higher
on predation treatments than under natural bird-cached
conditions because surface-sown seeds attracted rodents.
Tc test the hypothesis that surface-sown seeds acted as an
22
attractant, 100 seeds each on mineral and litter were
singly buried on sites accessible to rodents.
Five areas
of 20 seeds each were laid oiit for each seedbed condition
with seeds buried 2 to 4 cm deep in a I dm by I dm grid
/
pattern.
Seed disturbance was 24% on mineral and 40% on
litter seedbed.
Since disturbance was observed on 100% of
the buried seed locations in the predator exclusion
treatments allowing access to rodents (EB and EN), it
appears that surface-sown seeds did act as a rodent
attractant.
Seeds that germinated on the exclude birds only and
exclude none treatments were buried seeds that rodents
looked for but did not find.
There was no significant
difference in percent emergence of whitebark seedlings
between exclude birds only and exclude none treatments
within a year (Table 3).
There were, however, significant (p=0.056) between
year differences in the exclude birds only and exclude
none treatments.
Significant differences between years
for percent emergence could have been due to differences
in viability between seedlots, predator populations, or
climatic factors.
23
Table 3.
Emergency of whitebark pine for exclude birds
only and'exclude none treatments in 1988 and
1989.
Treatment
Exclude
birds
only
Year
Exclude
none
1988
2.1 a
2.5 a
1989
3.6 a
3.5 a
- Percent germination within year, sharing a common letter
are not significantly different (p-value<0.I).
Viability was 95% and 97% respectively for seed
planted in 1987 and 1988; thus the probability that
emergence differences were due to different seedlots was
low.
Seeds were x-rayed to ensure that only filled seeds
were planted.
It is also unlikely that year to year differences
were due to rodent effects.
Rodent populations did not
appear to vary enough to suggest that they caused
between-year emergence differences.
Rodent catches were
19 in October of 1987, and 16 in July and 11 in September
of 1988.
Precipitation was probably responsible for the
between-year differences in seedling emergence.
The study
site was clear of snow and soils were at field capacity
immediately after the first week of June in both
measurement years.
Emergence was noted from mid-June
24
through the first of September in both years.
Precipitation records for two Soil Conservation Service
weather stations,- Mill Creek to the north and Canyon to
the southeast of the study area were obtained to,examine
area wide moisture patterns (Tables 12 and 13, Appendix).
The weather stations demonstrated dry conditions (26% and
17% of normal), in 1988, and near normal (79% and 72% of
normal), in 1989, for the months of June through August.
The three to four fold difference in precipitation between
1988 and 1989 was distributed evenly throughout the spring
and summer months.
Emergence
There was no predation of surface sown or buried
whitebark pine seeds on exclude birds and rodents and
exclude rodents only treatments.
These treatments thus
provide an estimate of maximum emergence under field
conditions.
Analysis of variance was used to evaluate the
effects of biotic and microsite factors on whitebark pine
emergence.
Years
Analysis of variance showed that emergence varied
significantly between planting years (Table 4).
Therefore,
analysis of biotic and microsite factor affects on
emergence and survival of whitebark pine was conducted
25
Table 4.
Effect of year and biotic and microsite factors
on emergence of whitebark pine. An ANOVA.
Emergence was measured as a proportion and
transformed to the arc sine of the square root
of the proportion (Snedecor and Cochran 1980).
Factor
Year (1988-1989)
Predator exclusion
Shade cover
Seedbed condition
Sowing depth
Error
Total
df
I
I
2
2
I
172
179
within individual years.
SS
MS
1.3381
0.1094
0.0881
0.4831
5.6318
2.8586
10.5091
1.3381
0.1094
0.0440
0.2415
5.6318
0.0166
F
80.52
6.58
2.65
14.53
338.86
P
.0001
.0112
.0735
.0001
.0001
Between-year differences in
emergence were attributed to precipitation differences
(Tables 12 and 13, Appendix).
Seedbed and sowing depth significantly affected total
summer emergence in both years, while predator exclusion
was significant in 1988 only, and shade was significant in
1989 only. (Tables 5 and 6).
Significant differences in
emergence among replicates occurred in both measurement
years.
Significance of all factors varied by measurement
date as the summer progressed.
Tables 14 and 15 (Appendix)
show the 1988 and 1989 analysis of variance results for
biotic and microsite factors and two-way interactions as
the cumulative percent emergence of whitebark pine for
each recording date.
Significance of all factors and
interactions increased as germination progressed through
the summer months.
26
Table 5.
Effect- of biotic and microsite factors and their
interactions on 1988 emergence of whitebark
pine: An ANOVA. Emergence was measured as a
proportion and transformed to the arc sine of
the square root of the proportion (Snedecor and
Cochran 1980).
Factor
df
SS
Replicate
Predator exclusion
Shade cover
Seedbed condition
Sowing depth
Pred x Shade
Pred x Seed
Pred x Sow
Shade x Seed
Error
Total
2
I
2
I
2
2
I
I
2
58
71
0.2975
0.1648
0.0237
0.0692
0.7385
0.0330
0.0335
0.0530
0.0509
0.4824
1.9466
Table 6.
' MS
0.0487
0.1648
0.0119
0.0692
0.7385
0.0165
0.0335
0.0530
0.0254
0.0083
F
P
17.88
19.81
1.43
8.32
88.79
1.99
4.03
6.37
3.06
.0001
.0001
.2483
.0055
.0001
.1466
.0494
.0144
.0544
Effect of biotic and microsite factors and their
interactions on 1989 emergence of whitebark
pine: AN ANOVA. Emergence was measured as a
proportion and transformed to the arc sine of
the square root of the proportion (Snedecor and
Cochran 1980).
Factor
Replicate
Predator exclusion
Shade cover
Seedbed condition
Sowing depth
Pred x Shade
Pred x Seed
Seed x Sow
Error
Total
df
2
I
2
2
I
2
2
2
93
107
SS
0.0472
0.0091
0.0895
0.4718
5.5792
0.0327
0.0553
0.1221
0.8175
7.2244
MS
0.0236
0.0091
0.0447
0.2359
5.5792
0.0163
0.0276
0.0611
0.0088
F
2.68
1.04
5.09
26.84
634.73
1.86
3.14
6.95
P
.0736
.3110
.0080
.0001
.0001
.1618
.0477
.0015
27
Total numbers of whitebark emergents varied
significantly between years but the emergence curves were
similarly shaped (Figure 4).
SurfaCe-sown seed apparently
germinated about I week before buried seeds in both years
(Figure 4).
While seeds buried 2 to 4 cm may have
germinated (when radicle first extends through the seed
coat) at the same time as surface-sown seed, germination
could not be recorded until emergence occurred.
There was
no germination after the first of July for surface-sown
seeds; perhaps surface conditions were too dry to allow
germination or all nondormant seeds had germinated by
then.
After July emergence from buried seeds was
significantly higher than from surface-sown seeds in both
years (Figure 4).
Emergence from buried seeds was rapid
before the first of August and sparse thereafter.
Predator Exclusion
Whitebark pine emergence differed significantly
between the exclude birds and rodents (EA) and exclude
rodents (ER) only treatments in 1988 but not in 1989
(Tables 5 and 6).
Emergence differences were attributed
to screen design differences and not bird predation since
no seeds were removed from the pests.
The percent
emergence was 9.5% for EA and 3.8%-for ER treatments in
1988 (Table 7).
The higher emergence rate in the EA
treatment is attributed to the milder plot microclimate of
CUMULATIVE NUMBER OF EMERGENTS
900
-
800
-
700
-
600
-
500
-
400
-
Buried,
moist 1989
300 200
Buried,
-
Surface-sown,
100 -
dry 1988
moist 1989
S u r f a ce^ s own. dry 1988
AUGUST
Figure 4.
SEPTEMBER
OCTOBER
Cumulative number of whitebark pine emergents from buried and surface
sown seeds in 2 years.
29
Table .7.
Percent emergence of whitebark pine as affected
by predator exclusion, shade cover, seedbed
condition, sowing depth, and replicate on
treatments excluding rodents and birds (EA) and
excluding rodents only (ER) for 1988 and 1989
first-year and 1989 delayed emergents from the
1987 sowing. Values are not cumulative over
years.
Predator
exclusion
EA
ER
First--year
1989
1988
mean
mean
I / - percent
9.5 (a)-' 19.3 ( C )
17.9 (c)
3.8 (b)
Shade cover
(percent)
0
25
50
5.2 (a)
7.9 (a)
6.8 (a)
16.7 ( C )
18.8 (cd)
20.4 (d)
21.. 3 (e)
28.6 (f)
27.2 (f)
Seedbed
condition
Mineral
Litter
Burned
8.2 (a)
5.1 (b)
—
26.0 (c)
14.5 (d)
15.3 (d)
24.9 (e)
26.5 (e)
Sowing depth
Surface
Buried
1.8 (a)
11.5 (b)
3.5 ( C )
33.7 (d)
6.0 (e)
45.4 (f)
7.9 (a)
9.5 (a)
2.5 (b)
19.3 (c)
20.2 (c)
16.4 (d.)
20.8 (e)
29.4 (f)
26.9 (f)
Factor
Factor
level
Replicate
I
2
3
Delayed
1989
mean
20.9 (e)
30.5 (f)
I /Similar and dissimilar letters in parentheses within a
column for a factor represent nonsignificant and
significant differences respectively.
the EA treatment.
The EA treatment has hardware cloth,
with 0.63 cm-square holes, about 4 inches above the ground
level and totally enclosing the plot.
The metal cloth may
have provided extra shade, reduced daytime temperatures,
and increased night temperatures.
Screening method slightly increased emergence of
whitebark pine in moist 1989 (Table 6).
I attribute the
between-year difference to the increased summer rains in
Jl
30
1989 (Tables 12 and 13, Appendix):
emergence was enhanced
by screening affects (cooling or reducing evaporation)
provided by the exclude birds and rodents, treatments in
dry years.
Screening effects disappeared in 1989, when
moisture was not as limiting throughout the growing
season, as it was in 1988.
Shade Cover
In 1989 when water was less limiting, shade
significantly (p=0.008) affected emergence (Table 6).
Percent emergence of whitebark pine in 1989 was
significantly higher (20.4%), under 50% shade than with no
shade (16.7%) (Table 7).
The 25% shade treatment was
intermediate (18.8) in percent emergence and did not
significantly differ from 0% or 50% shade cover.
Surprisingly shade did not significantly improve emergence
of whitebark pine seed in dry 1988 (p= 0.2483) (Table 6).
The 1988 emergence of whitebark pine appeared higher for
shaded than nonshaded treatments, ranging from 7.9%, 6.8%,
to 5.2% emergence with 25%, 50%, and 0% shade cover,
respectively (Table 7).
The snow fence slats caused
alternate strips of shade and full sunlight.
Dead shade
may not have been enough to override the effect of drought
on emergence in a dry year, even under 50% shade.
I
31
Seedbed Condition
Seedbed condition significantly affected whitebark
. pine germination in both ybars (Tables- 5 and 6) .
Emergence
of whitebark pine was significantly higher on mineral
seedbeds than on litter, or burned seedbeds in both years.
There was no burned seedbed treatment in 1988.
Most conifers germinate best on mineral seedbeds
(Schmidt and Lotan 1980; Seidel 1979; Zasada et al. 1978).
The exposure of soil is expected to eliminate competing
vegetation, and provide more light, moisture, and nutrients
. for seedling growth (Schmidt et al. 1976).
Although no
quantitative measures of competing vegetation were taken,
reduction of competition may have been one of the primary
reasons for better whitebark emergence on mineral seedbeds.
Moisture may have been the factor limiting emergence on
litter seedbeds on these high elevation sites.
Soil
moisture was consistently higher on litter than mineral
seedbeds except in the fall (a season of low mortality
(Figure 5).
A significantly higher percent of whitebark germinants
emerged on mineral (26.0) than on litter (14.5) or burned
(15.3) seedbeds in 1989 (Table 7).
I expected emergence
on the burned seedbeds to be as high as on mineral soil
due to the nutrient flush following burning and the
reduction in competing vegetation.
My burn treatments may
have been too light.to fully provide these benefits.
1988
OX Shade Cover
SOIL WATER (X)
♦-- * LITTEH
•-- • MINERAL
AUGUST
♦— ♦ LITTER
•--- • MINERAL
Figure 5.
1988
50% Shade Cover
♦---* LITTER
•---• MINERAL
AUGUST1 SEPT
1989
0% Shade Cover
AUGUST
1988
25% Shade Cover
AUGUST1 SEPT
1989
25% Shade Cover
♦
•
*--- - LITTER
----• MINERAL
1989
50% Shade Cover
------- ------- LITTER
------- • MINERAL
♦
•
OJ
to
------- ------- LITTER
------- • MINERAL
AUGUST
Soil water in top 5 cm of soil on mineral and litter seedbeds under
0%, 25% and 50% shade cover in 2 years.
33
Emergence on burned seedbeds in subalpine forests is
usually comparable to that on litter initially, it
increases rapidly with time, and eventually surpasses
emergence on mineral (scarified) seedbeds (Fiedler 1980).
Emergence of whitebark on my burned treatments may
eventually equal that on mineral seedbeds.
Light burning
intensities do not have as severe effects on organic
matter and soil properties as more intense burns.
Hot
fires consume most of the organic material, alter soil
structure, and create unfavorable conditions for seedling
growth (Vogl and Ryder 1969).
Hot fires also volatilize
nitrogen and result in loss of other nutrients in fly ash
(Perry 1979).
Sowing Depth
Even in the absence of predation (EB and ER; discussed
above) deep sowing significantly improved emergence of
whitebark pine in both measurement years (Tables 5 and 6).
Emergence of buried seed was 11.5% and 33.7% compared to
1.8% and 3.5% for surface-sown seed in 1988 and 1989
respectively (Table 7).
As previously discussed emergence
was significantly lower in dry 1988 for both buried and
surface-sown seeds.
The Clark's nutcracker caches whitebark pine seed at
a depth of 2 to 4 cm on ground surfaces ranging from
mineral, to litter, and gravel (banner 1980).
Some
( i) IL
34
factors which may reduce surface germination are lower
water availability and more exposure to solar radiation
(Gorski and Gorska 1979) .
Replicate
Whitebark pine emergence was significantly less in
replicate 3 than in replicates I or 2 in both 1988 and
1989.
Percent emergence for each replicate was 7.9 and
19.3 for replicate I, 9.5 and 20.2 for replicate 2, and
2.5 and 16.4 for replicate 3 for 1988 and 1989 respectively
(Table 7).
The within-year difference may be attributable
to soil differences in the study area.
Soils were usually
drier in replicate 3 than in replicates I and 2 during the
early summer when emergence was greatest (Figure 6).
The soil on replicate 3 was identified as a Typic
Cryorthent, sandy skeletal with a 6-inch A horizon
containing 54% sand over a C horizon of 60% sand.
Soils
on the other replicates had a narrow 4 to 6 cm thick
cambic B horizon between the A and C horizons which
classified them as inceptisols.
The deficiency of moisture
likely limiting the emergence of whitebark pine on
replicate 3 is probably due to the high sand content of
the soil profile and the shallow A horizon.
The A and
B horizons on replicates I and 2 had less sand and more ■
silts and clays according to hand textural analysis.
50
1988
■<
+•
40 -
■ REPLICATE I
* REPLICATE
2
30 -
20
SOIL WATER (%)
-
10-
O
JUNE
I
JULY
I
AUGUST
I SEPTEMBER I
OCTOBER
REPLICATE I
REPLICATE 2
REPLICATE 3
Ground
Frozen
After
Mid-Sept.
AUGUST
Figure 6.
I SEPTEMBER 1 OCTOBER
Percent moisture in top 5 cm of mineral soil by
replicate for 1988 and 1989.
36
Soils on replicates I and 2 would have, a higher
water-holding capacity than on the sandy soils of
replicate 3.
Interactions
Four two-factor interactions of biotic and microsite
factors in 1988 and three two-way interactions in 1989
appeared to affect whitebark pine emergence (Tables 5 and
6).
No three-way or higher interactions were significant.
1988 Interactions.
The 1988 interactions and their
significance on emergence of whitebark were:
predator
exclusion by percent shade cover (A) (p=.1466), seedbed
condition (B) (p=.0494), and sowing depth (C) (p=.0144),
and percent shade cover by seedbed condition (D) (p=.0544)
(Table 5).
The interaction of predator exclusion by
percent shade cover showed relative differences in mean
percent emergence between the 0 (6.3%), 25 (12.1%), and 50
(9.4%) percent shade cover on the exclude birds and
rodents (BA) treatment (Figure 7A). A p-value of .1466 is
not considered significant at the p<0.I level but is low
enough to indicate that a relationship may exist.
There
was no difference in mean percent emergence between the
three shade levels on the exclude rodents only (ER)
treatment.
Shade level
CZ=IOX
EZZl 25%
CZZ 50%
L U 10
PREDATOR EXCLUSION
I-- 1Surface-sown
PREDATOR EXCLUSION
I Z - J Mineral
EZlLitter
25
PREDATOR EXCLUSION
Figure 7.
PERCENT SHADE COVER
Two-factor interactions of the 1988 biotic and microsite factors
affecting whitebark pine emergence: predator exclusion by shade level
(A), seedbed condition (B), sowing depth (C), and percent shade cover
by seedbed (D).
38
Emergence increased considerably when plots were
shaded by the EA screen exclosures and shade covers.
The
change in total numbers may be caused by microclimate
differences between predator exclusion treatments due to
effects of metal screen exclosures.
The interaction of predator exclusion by seedbed
condition showed differences in mean percent emergence of
whitebark pine between mineral and litter seedbed under
the EA treatment (Figure 7B).
The emergence differences
ameliorated with the ER treatment.
I attribute seedbed
treatment differences to microclimate modification from
the screening on exclude birds and rodents treatments.
Emergence of buried whitebark seed was greater than
surface sown seed and this difference was much larger on
EA than on ER treatments (Figure 1C).
Emergence
differences between predation treatments may again be
caused by microclimate effects on EA treatments.
While shading did not (p=0.2483) benefit whitebark
pine emergence overall, it did affect it on mineral
seedbeds (Figure 7D).
Emergence of whitebark on litter
seedbeds remained constant under the three shade levels
while emergence on mineral seedbeds increased considerably
under 25% but only slightly under 50% shade over nonshaded
conditions.
Emergence between seedbeds was not different
on nonshaded conditions but was higher on mineral than on
litter under 25% and 50%.
Because shading had a negative
39
effect on soil moisture in the top 5 cm of soil (Figure 5)
I tentatively attribute its positive effects on emergence
to reduction of surface temperature on mineral soils„
1989 Interactions.
The three two-factor interactions
affecting 1989 emergence of whitebark pine are predator
exclusion by shade cover (A) (p=0.1618) and seedbed
condition (B) (p=0.0477) and seedbed condition by sowing
depth (C) (p=0.0015)
(Figure 8).
Similar to 1988, there
were differences between the three shade levels on exclude
birds and rodents but not on exclude rodents only
treatments (Figure BA).
Explanations were previously
provided for the same 1988 interaction for between- and
within-treatment differences.
The interaction of predator exclusion by seedbed
condition was significant in both years.
In 1989, however,
a burned seedbed treatment was added. Again there was a
high mean difference in the within-treatment emergence of
whitebark pine between mineral and litter on exclude birds
and rodents (EA) treatments in 1989 (Figure SB).
Increased
precipitation is probably why emergence was. higher on
mineral seedbeds on both EA and ER treatments in 1989.
Although there was no significant effect of seedbed
in 1988, emergence differences due to seedbed preparation
increased slightly on the exclude rodents only (ER)
treatments in 1989 (Figures 7B and SB). Whitebark
emergence on burned and litter treatments remained low in
MEAN PERCENT EMERGENCE
LU
U
PREDATOR EXCLUSION
28
24
□
Mineral
EBLitter
C Z 2 Burn
20
16
12
8
4-
0
PREDATOR EXCLUSION
Figure 8.
Two-factor interactions of the 1989 biotic and microsite factors
affecting whitebark emergence: predator exclusion by shade level (A),
seedbed condition (B), and seedbed condition by sowing depth (C).
41
1989 on both EA and ER treatments.
There was no difference
in emergence between predation treatments in 1989 when
precipitation was near normal and evenly distributed
throughout the growing season.
Seedbed condition by sowing depth was a significant
interaction of emergence of whitebark pine in 1989.
Seedbed condition significantly affected emergence of
buried whitebark seeds and had no affect on surface sown
seeds.
Buried whitebark seeds had higher emergence than
did surface sown seeds on mineral, litter, and burned
seedbeds (Figure SC).
Germination of surface-sown seeds
was less than 4% on mineral seedbeds, decreasing slightly
from mineral to litter (2%) to burned seedbeds (1.5%).
Emergence of buried seeds was 26% on mineral seedbeds and
decreased to near 15% for seeds buried on litter and
burned seedbeds.
Delayed Emergence
I call germination in the second or third year
"delayed emergence".
Here, delayed emergence is the ratio
of 1989 emergents to the number of seeds that did not
germinate in 1988.
Plots seeded in 1987 were examined for
emergence in 1988 and again in 1989 to measure whitebark
pine delayed emergence.
Germination of the European
42
equivalent of whitebark pine (Pinus cembra)(Critchfield
and Little 1966) may be delayed until the second or even
third year after dispersal (Krugman and Jenkinson 1974)
More of the seeds sown in 1987 emerged in 1989 than
in 1988 (Table 7).
Of seeds sown in 1987 the 14.6%
emergence in the second summer was significantly more than
first-year emergence in 1988 (4.5%) and in 1989 (10.2%).
The high dormancy rate of 1987 sown seeds in 1988 may
have been due to drought enforcement.
High delayed
emergence of 1987 sown seed (higher in relative terms than
first-year emergence in 1989) was observed under most main
factor levels.
The higher delayed emergence of whitebark over
first-year emergence may be due to impermeable seed coat,
the presence of physiological embryo dormancy (Pitel and
Wang 1980), or a combination of these and other unknown
factors.
Physiological embryo dormancy is usually overcome
by certain metabolic events resulting in decreased
inhibitor and increased growth promoter content, increased
energy charge, and derepression and activation of the
genome and increased protein synthesis (Kahn 1977).
study does not evaluate specific reasons for delayed
emergence.
This
43
Table 8.
Effects of biotic and microsite factors and
significant interactions on delayed emergence of
whitebark pine: An ANOVA. Emergence was measured
as a proportion and transformed to the arc sine
of the square root of the proportion (Snedecor
and Cochran 11)80)..
Factor
Replicate
Predator exclusion
Shade
Seedbed condition
Sowing depth
Pred x Shade
Pred x Sow
Bed x Sow
Error
Total
df
2
I
2
I
I
2
I
I
. 60
71
SS
0.1054
0.1666
0.1439
0.0112
4.9561
0.1033
0.3514
0.4317
1.1043
7.3738
MS
0.0527
0.1666
0.0720
0.0112
4.9561
0.0517
0.3514
0.4317
0.0184
F
2.86
9.05
3.91
0.61
269.29
2.81
19.09
23.45
P
.0650
.0038
.0253
.4380
.0001
.0683
.0001
.0001
Delayed whitebark emergence varied significantly
between predator exclusion treatments, shade cover, sowing
depth, and replicate (Table 8).
Seedbed condition did not
affect delayed emergence as it did first-year emergence in
both 1988 and 1989.
To determine the natural occurrence of this phenomena
I looked for delayed emergence in the field.
Seedling
clusters on the study site were dug up and intact seeds
were taken to the laboratory.
X-ray analysis showed that
75% (12 seeds) of the intact non-germinated seeds had
fully developed endosperms.
The filled seeds were
cold-moist stratified for 30 days (simulating a second
year of natural stratification) and put in germination
chambers under temperature and moisture conditions
recommended by Jacobs and Weaver (1990) .
Germination was
44
75% for the filled seeds.
Rates of delayed whitebark
germination appear to be higher than for first-year
germination.
Predator Exclusion
Screen design significantly affected delayed
germination and emergence as it affected first-year
emergence in both 1988 and 1989.
Delayed emergence on
exclude rodents only treatments (30.5%) was significantly
higher than on the exclude birds and rodents treatment
(20.9%), the opposite of first-year emergence (Table 7).
Delayed emergence rates for both predator treatments were
expected to be similar to 1989 first-year emergence
because of the moderating effect of higher 1989
precipitation.
I can offer no reason why the effect of
predator exclusion on emergence in 1988 was reversed for
the 1989 delayed emergents.
Shade Cover
Whitebark emergence was significantly higher for
shaded than nonshaded treatments for both delayed (Table 8)
and first-year emergence in 1989 (Table 7).
Increased
precipitation in 1989 probably reduced the effect of
summer drought on germination, enhancing the effect of
shade.
45
Sowing Depth
Delayed emergence was significantly higher than
first-year emergence in both 1988 and 1989 (Table 8) for
surface-sown and buried whitebark seeds.
Delayed emergence
of buried seeds was significantly higher (45.4%) than
surface sown seeds (6%) (Table 7);.
Replicate
Replicate I had significantly less delayed emergence
than replicates 2 and 3 (Table 7).
While lower first-year
emergence in replicate 3 may be explained by textural and
soil horizon differences as previously described, I cannot
explain why delayed emergence is lowest on replicate I.
Interactions
Three two-factor interactions of biotic and microsite
factors affected delayed emergence of whitebark pine in
1989 (Table 8); predator exclusion by shade level (B)
(p=0.0683) and sowing depth (B) (p=0.0001) and seedbed
condition by sowing depth (C) (p=0.0001)
(Figure 9).
The
number of delayed emergents was higher on ER than EA
treatments for 0% and 25% shade cover and did not change
between treatments for 50% shade cover (Figure 9A) .
Higher delayed emergence on ER treatments was reversed
from first-year emergence in 1988 when EA treatments had
higher emergence for all three shade treatments.
BO
50-
MEAN PERCENT EMERGENCE
40 -
Shade level
CZ=IOX
CZZESX
CZZl50X
A
LU
O
3020-
100-
Figure 9.
PREDATOR EXCLUSION
<T\
Mineral
LU
Litter
SEEDBED CONDITION
Two-factor interactions of the 1989 biotic and microsite factors
affecting delayed emergence of whitebark pine: predator exclusion by
shade level (A), and sowing depth (B) and seedbed condition by sowing
depth (C).
47
First-year emergence in 1989 was the same for 0% and 25%
and decreased for 50% shade cover between EA and ER
treatments.
Precipitation could have caused the
between-year differences in emergence but I do not know
why 1989 emergence patterns for delayed and first-year
emergents were different.
Delayed emergence of whitebark varied significantly
within predator exclusion treatment for both buried and
surface-sown seeds and between exclusion treatments for
buried seeds (Figure 9B).
In 1988 first-year emergence
was higher on EA treatments.
Delayed emergence of buried
seeds was significantly higher on ER than EA treatments
because more seeds had enforced dormancy in 1988 on the ER
treatment.
Buried whitebark seeds had more delayed emergence
than surface-sown seeds on mineral and litter seedbeds
(Figure 9C).
Delayed emergence was 10.2% and 39.7% on
mineral and 1.9% and 51.3% on litter seedbeds for
surface-sown and buried seeds respectively.
Delayed
emergence of surface-sown seeds was lower on litter than
on mineral seedbeds but the relationship was reversed for
buried seeds.
The increase in delayed emergence for
buried seeds and decrease in surface-sown seeds could be
due to a measurement error.
The 1987 surface-sown seeds
were positioned close to buried seeds (Figure 3).
and.summer rains moved and partially buried many
Winter
48
surface-sown seeds.
When seeds emerged it was difficult
to distinguish if they had originally been a surface-sown
or a buried seed.
Sowing depth treatments were separated
in 1988 to eliminate or reduce the overlapping of sowing
types (Figure 3).
It was impossible to stop the natural •
burying of surface-sown seeds by snow, water, and winds.
Mortality
Whitebark pine emergence rates were high from mid-June
through the end of July, slowed, and were nil in August
(Figure 4).
Mortality of first-year whitebark seedlings
followed the same pattern with the highest mortality rates
occurring when emergence was highest (Figure 10).
rates decreased after the first of August.
Both
Through the
first of August emergence was always greater than mortality
so there was a net accumulation of surviving seedlings.
Germinants continued to emerge from early August until the
first of September but the total numbers of survivors
remained constant because mortality rates equaled emergence
rates.
Mortality began near the end of June and continued at
a low level until early September (Figure 10).
causes of mortality were identified:
Three
(I) insolation (heat
scorching of seedling stem at ground surface), (2) drought
49
50
1988
I *
B a
□
40
INSOLATION
DROUGHT
ANIMAL
30
NUMBER OF DEATHS
20
10
. I
0
160
Figure 10.
167
174
188
41--PL 1
195
202
208
JUNE
I
JULIAN DATE
JULY
JUNE
I
JULIAN DATE
JULY
—
216
0— 1
230
243
254
292
I
AUGUST
I SEPT. I
I
AUGUST
I SEPT. I
First-year mortality of whitebark pine
seedlings by cause over time.
50
(drying out of seedling), and (3) animal (burial,
uprooting, or nipping).
Fungi and insect caused mortality
were never seen on whitebark seedlings.
Insolation mortality of 1988 emergents began in late
June when day lengths were the longest and ended by early
August (Figure 10).
Insolation mortality slowed during
late July and ended around mid-August in 1988 and 1989.
Drought mortality occurred later in the growing
season than insolation— except in early July of 1989 when
higher numbers of whitebark emergents died from drought
than insolation (Figure 10B).
Most of the first emergents
dying from drought were surface germinated seeds.
This
was probably due to the fact that roots of surface-sown
seeds have a greater distance to reach the more dependable
water supplies of mineral soil.
Drought mortality of
seedlings germinated from buried whitebark seeds began
around the third week in July and continued until the
first of September.
The only animal caused mortality observed was in 1989
when 11 seedlings were buried by soil brought to the
surface by pocket gopher activity.
There was no evidence
of pocket gophers feeding on seeds or seedlings.
Increased survival of buried seeds compared to
surface-sown seeds was evident in both years and more so
in moist 1989 than in dry 1988 (Figure'll).
Only
surface-sown seeds lying on favorable microsites germinated
51
Buried
SURVIVAL (%)
Surface-sown
AUGUST
I SEPTEMBER I OCTOBER
Buried
Surface-sown
AUGUST
Figure 11.
SEPTEMBER ' OCTOBER
Survival of whitebark pine seedlings germinated
from buried and surface sown seeds.
52
Table 9.
Mortality of 1988 and 1989 whitebark pine
emergents. on mineral and litter seedbeds under
0%, 25%, and 50% shade. Mortality is the
percent of emergents in each category that died.
Year
Percent
shade
Seedbed
condition
1988
0
Mineral
Litter
25
Mineral
Litter
6
0
8
29
0
0
14
29
50
Mineral
Litter
5
9
12
22
0
0
17
31
0
Mineral
Litter
Burned
I
10
14
2
4
6
0
0
0
3
14
19
25
Mineral
Litter
Burned
3
2
6■
5
2
4
0
0
0
8
4
10
50
Mineral
Litter
Burned
I
0
0
4
17
7
6
0
0
11
17
7
Insolation Drought Animal Total
- percent
33
11
0
44
23
6 •
0
29
eee
1989
p_
when moisture was limiting in 1988 and thus many survived.
The high precipitation of 1989 resulted in a four-fold
increase over 1988 in the number of emergents from
surface-sown seed.
Many surface-sown seeds germinated in
1989 because of increased precipitation, but because of
unfavorable microsites most may have succumbed to drought.
Shade helped reduce seedling mortality in 1988 but
not in 1989 (Table 9).
And shading influenced the type of
mortality on mineral, litter, and burned seedbeds.
Insolation mortality of seedlings was highest on seedbeds
with no shade cover in 1988; shading of as little as 25%
53
Table 10.
Maximum surface temperatures recorded on
mineral, litter., and burned seedbeds under 0%,
25%, and 50% shade cover in 1988 and 1989.
There was no burned treatment in 1988.
Seedbed
condition
No. of
obs.
1988
Mineral
Litter
18
18
65
73
59
65
59, .
73-/
1989
Mineral
Litter
Burned
27
27
27
59
73
79
52
65
65
52
65
73
Year
y
Percent shade cover
0
25
50
This high temperature was an anomaly because wax
temperature pellets were exposed to the sun.
decreased insolation mortality (Table 9).
Insolation
mortality on mineral seedbeds was higher than for litter
seedbeds despite higher surface temperatures on litter
(Table 10).
In 1989, insolation mortality of whitebark pine
seedlings was highest on burned (14), second highest on
litter (10), and lowest on mineral seedbeds (I) in the
absence of shade cover (Table 9).
Shade reduced surface
temperatures (Table 10) and decreased seedling mortality
due to insolation.
Insolation mortality averaged 8.3%,
3.6%, and 0.3% under 0%, 25%, and 50% shade respectively
(Table 9).
Drought mortality of whitebark pine seedlings was
lowest on nonshaded plots in 1988.
It was lower on both
0% and 25% shade covered plots in 1989 (Table 9).
As
shade increased drought mortality increased, perhaps
IL
' 54
because shading reduced soil moisture under shade coverings
by deflecting rain (Figures 12 and 13).
This is consistent
with the fact that nonshaded plots were moister than
shaded plots during the mid-summer weeks when most drought
mortality was occurring.
No deflection of rain by shade
covers may account for higher soil moisture values on open
plots.
On non-predation plots (BA and ER) at the end of the
second growing season, 81% of the whitebark pine seedlings
that had germinated and survived the first growing season
were still alive.
There were nearly five times more
seedlings on BA and ER plots than on EB and EN plots;
these seedlings were protected, however, from animal
predation thus giving unnaturally low mortality results.
A more realistic picture of natural mortality is
obtained from EB and EN treatments allowing both the
effects of climate and predation on seedlings.
Mortality
counts for whitebark pine seedlings were taken in the
spring and fall of 1989.
Nearly 58% of the seedlings
survived the first winter on EB and EN treatments.
There
was no mortality on these treatments during the 1989
growing season.
Of the overwinter mortality, the causes
for whitebark seedling loss were:
insolation - (0%),
drought - (0%), animal - (6%), and unknown - (94%).
One
seedling was nipped by rodents and five were buried by
55
SOIL WATER (%)
Mineral Soil
•- - - - - ■ OX Shade
+. . . . . + 25% Shade
♦
- o 50% Shade
August
Lit ter- C o v e r e d Soil
August
Figure 12.
1September 1 October
■- - - - - ■ OX Shade
+.... + 25% Shade
♦---- ♦ 50% Shade
1September 1 October
Percent moisture in top 5 cm of soil on
mineral and litter seedbeds under 0%, 25%, and
50% shade in 1988.
56
SOIL WATER (%)
Mineral Soil
■---- - OX Shade
+. . . . + 25% Shade
♦- - - - ♦ 50% Shade
August
50
.---- . 0% Shade
+. . . . + 25% Shade
♦- - - - ♦ 50% Shade
Litter-Covered Soil
40
V\
30
20
10
0
June
Figure 13.
July
*
August
I Sept.
Percent moisture in top 5 cm of soil on
mineral and litter seedbeds under 0%, 25%, and
50% shade in 1989.
57
pocket gophers„
Causes of winter mortality could not be
determined because the area is covered by approximately 60
to 150 cm of snow.
Subsurface Soil Temperatures
Germination of seeds depends largely on soil
temperature and moisture conditions (Fitter and Hay 1981).
Germination of whitebark occurs between 10° and 40°C, has
an optimum range for germination enhancement from 15° to
35 °C, and peaks at 35 °C (Jacobs and Weaver 1990) .
Moisture
is not a limiting factor in laboratory studies; however,
when soil temperatures in the field reach 35°C, moisture
probably becomes the factor limiting germination.
Minimum subsurface temperatures were generally warmer
with increased shade cover, indicating a moderating effect
of treatment on soil temperature (Table 11).
In general,
mineral seedbeds generally had relatively low minimum
temperatures and high maximum temperatures; litter mulch
moderated soil temperature swings.
Temperature variation
between shade levels was high for maximum and low for
minimum subsurface temperatures throughout the growing
season (Figures'14 and 15).
Maximum temperatures of
around 38°C occurred near the first of August when soil
moisture was approximately 50% of field capacity
(Figures 12 and 13).
58
Table 11.
Minimum-maximum soil temperatures at a depth of
2.5 cm on mineral and litter seedbeds under 0%,
25%, and 50% shade cover— 1988 and 1989.
Values in parentheses are mean minima and
maxima for the summer
Year
Seedbed
condition
1988
Mineral
Minimum
Maximum
Percent shade cover
0
25
50
- - 0C -12 ( I) - 8 ( 3) - 7 ( I)
37 (26)
28 (20)
29(23)
Litter
Minimum
Maximum
- 9( 2)
27(23)
- 8 ( 2)
24 (19)
- 7 ( 2)
28 (21)
Mineral
Minimum
Maximum
-12 (-2)
36(25)
- 8 (-1)
26(22)
- 9 (-1)
29(23)
Litter
Minimum
Maximum
- 9 (-1)
29(23)
- 8 (-1)
30 (20)
-11( 0)
23(17)
1989
Temperature
59
TEMPERATURE (DEGREES C)
MINERAL
-10
-
-- OX Shade
.. 25% Shade
-- 50% Shade
Extremes
August
September
LITTER
-10
-
-..
— —
—
0% Shade
25% Shade
50% Shade
Extremes
August
Figure 14.
September
Mean minimum-maximum soil temperatures at a
depth of 2.5 cm on mineral and litter seedbeds
under 0%, 25%, and 50% shade cover by date for
1988. Bold lines represent the minimum and
maximum temperatures recorded.
60
TEMPERATURE (DEGREES C)
MINERAL
-- OX Shade
.. 25% Shade
-- 50% Shade
—
Extremes
August
September
LITTER
“ 10
-
-- 0% Shade
.. 25% Shade
— — 50% Shade
Extremes
August
Figure 15.
September
Mean minimum-maximum soil temperatures at a
depth of 2.5 cm on mineral and litter seedbeds
under 0%, 25%, and 50% shade cover by date for
1989. Bold lines represent the minimum and
maximum temperatures recorded.
61
CHAPTER 4
CONCLUSIONS
Seed survival, germination, and seedling survival of
whitebark pine are affected by biotic and microsite
factors including predation, shade cover, seedbed, and
sowing depth.
Seed Losses
Evaluation of seed losses from predation indicate
that:
1.
Where it is accessible to them rodents removed
essentially all surface-sown and most buried seeds.
Survivors germinate the first and second year after
planting.
Mammals would probably have found fewer buried
seeds if nearby surface-sown seeds had not acted as an
attractant and, if so, the number of emergents would have
been greater.
2.
Avian predators did not remove whitebark seeds exposed
on the ground.
Seed security must have resulted from the
inability of the birds to detect the seeds.
Clark's
62
nutcrackers were observed sitting on several exclosures,
but none were seen foraging for seed inside exclosures
open to birds.
Emergence
Emergence of first-year and delayed whitebark pine is
affected by sowing depth, seedbed condition, and shade
cover, in decreasing order of significance.
The collective
effects of those factors were different for seeds that
germinated the first summer after seeding than for those
that carried over germinated in the second summer (delayed
emergence).
I.
Results from this study indicate that:
Probably due to microclimate effects, emergence rate
of whitebark seeds is higher on treatments that exclude
birds and rodents than on exclude rodents only treatments.
Both these treatments protected all sown seeds from
predation.
The overtopping screen likely modified the
microsite climate on treatments that exclude birds and
rodents by retaining heat during the nights and reducing
temperatures during the day.
I
63
2.
More sound whitebark pine seed germinated in the
second (delayed emergence) than in the first spring after
planting.
Emergence differences between delayed and
first-year emergence within a year are attributed to a
dormancy delay.
3.
First-year emergence was higher in moist 1989 than in
dry 1988.
The low germination of 1988 is attributed to
drought enforcement.
4.
When predation is eliminated emergence of buried seeds
is significantly higher than that of surface—sown seed.
In the field this effect will be magnified because few
surface-sown seeds escape mammal predation.
5.
Emergence of surface-sown seeds occurs immediately
after snow melt and ends by late June.
Buried seeds begin
emergence later (late June) and emergence is complete by
the first of August.
Most emergence is completed before
optimum soil temperatures are reached in early August.
6.
Increased shade cover increases emergence of whitebark
pine seedlings.
First-year emergence of whitebark is
slightly higher on shaded plots under low precipitation
64
conditions (1988) and significantly higher for both
first-year and delayed emergence when precipitation is
near normal (1989).
7.
First-year emergence is not different between seedbed
types under non-shaded conditions, but it is significantly
higher on shaded-mineral than on shaded-litter seedbeds.
There was no interaction of shade cover and seedbed
condition for delayed emergents.
8.
First-year emergence was affected by soil conditions.
Emergence was lower on sandy soils with only a thin
A horizon than on soils with both A and B horizons.
Delayed emergence was not affected by seedbed condition.
Mortality
Mortality of whitebark pine emergents switches from
insolation to drought to animal causation as the year
progresses.
I.
Mortality results indicate that:
On nonpredation plots seedling mortality during the
first growing season was due to drought (10% and 6%),
insolation (13% and 3%), and animals (0% and 1%) for 1988
and 1989 respectively.
Animals caused mortality by
burying seedlings with soil brought to the surface by
pocket gophers.
r
65
2.
Shade cover generally decreased insolation mortality.
Insolation mortality begins in late June when sun angle is
highest, peaks in July and ceased in early August.
3.
Surface germinated seedlings usually died from drought
before the start of insolation mortality.
Radicle
extension is apparently too slow to ensure establishment
unless the seed are located on ideal microsites.
4.
Increasing shade cover increased drought mortality.
Shade covers may intercept precipitation, a critical
resource in dry 1988.
5.
Survival of surface-sown seeds is lower than :for
buried seeds with most mortality occurring immediately
after germination.
6.
Most first-year mortality occurs during the winter
months but, due to winter inaccessibility of the site, its
cause is unknown.
This study provides information on the regeneration
process of whitebark pine that will help forest managers
and researchers.
For example, we now know how much seed
loss is due to predation; when and how much emergence
occurs in dry and near normal precipitation years; how
66
much shade, which seedbed, and what sowing depth is best
for whitebark emergence; that whitebark seeds can delay
emergence for at least 2 years; and that several mortality
factors affect first-year seedling survival.
Several questions remain: Which habitat type-aspectelevational combinations are best for whitebark
germination, survival, and growth?
What are the seed
transfer limits for whitebark; longitudinally,
latitudinally, and eleyational?
Will the effect of
burning on whitebark emergence be more beneficial over
time as it is for other subalpine species (Fiedler.1980)?
Will whitebark germinate after 3 or more years of dormancy?
Results from this study are preliminary but begin to
explain seed germination and seedling survival
characteristics of whitebark pine and also point out the
need for additional information.
Even though these
results are not complete we now know that we can
artificially regenerate whitebark pine by placing seeds
5 cm deep in mineral soil seedbeds.
Fall planting will
provide natural stratification conditions.
We also know
that to be more successful we should provide 25% to 50%
shade to aid seed germination and seedling survival.
Shelterwood cuttings could be used to provide these shade
conditions.
Because seed-eating rodents are attracted to
concentrations of seeds, we should sow seeds at least I m
apart and that we should plant at least 15 times more
67
seeds than the number of trees we eventually want
established on the area.
We should wait at least 2 years
before accessing regeneration results from first-year and
delayed emergence.
It should be recognized that these
results are preliminary, that more results will be
forthcoming, and that these recommendations may change to
reflect future results.
68
LITERATURE CITED
69
Amman, G . D., "Characteristics of mountain pine beetles
reared in four pine hosts." Environmental Entomology,
11(3) (1982): 590-593.
Arno, S., "Whitebark pine cone crops— a diminishing source
of wildlife food?" Western Journal of Applied
Forestry, 1(3) (1986): 92-94.
Arno, S . and Hoff, R. J., "Silvics of whitebark pine
(Pinu,s albicaulis)." General Technical Report INT-253.
Ogden, UT: U.S. Department of Agriculture, Forest
Service, Intermountain Research Station, 1989. 11 p.
Arno, S . and. Weaver, T., "Whitebark pine community types
and their patterns on the landscape." In:
Proceedings— Whitebark pine ecosystems: ecology and
management of a high-mountain resource. Ogden, UT:
U.S. Department of Agriculture, Forest Service,
Intermountain Research Station, in press, 1990.
Borzan, Z., "The application of x-rays in the analysis of
seeds of forest species." Sumarski List, Zagreb
No. 5-6 (1973): 166-189. Translated by. Liliane Neji
ed. Cathy Wolfheim for NSF and USDA Forest Service,
Washington, DC (1977).
Craighead, J . J., Scaggs, G. B., and Sumner, J. S., "A
definitive system for analysis for grizzly bear
habitat and other wilderness resources." Wildlands
Institute Monograph I. Missoula, MT: University of
Montana, 1982. 279 p.
Critchfield, W. B., and Little, E . L., "Geographic
distribution of the pines of the world."
Miscellaneous Publication 991. Washington, DC: U.S.
Department of Agriculture, Forest Service, 1966.
97 p.
Eggers, D . E., "Management of whitebark pine as potential
grizzly bear habitat." In: Proceedings— grizzly bear
habitat symposium; 1985 April 30-May 2; Missoula, MT.
General Technical Report INT-207. Ogden, UT: U.S.
Department of Agriculture, Forest Service,
Intermountain Research Station, (1986): 170-175.
Fiedler, C. E., "Analysis of regeneration in the subalpine
fir zone of western'Montana." Manuscript on file.at:
U.S. Department of Agriculture, Forest Service,
Intermountain Research Station, Forestry Sciences
Laboratory, Bozeman, MT, 1980. 64 p.
70
Fitter, A. H., and Hay, R„ K . M „ , "Environmental physiology
of plants." New York, NY: Academic Press, 1983.
355 p.
Forcella, F., "Flora, chorology, biomass and productivity
of the Pinus aIbicaulis-Vaccinium scoparium
association." Madrono 25 (1978) : 139-150.
Hutchins, H . E., Grand Rapids, MN: Itasca Community
College, Department of Forestry, personal
communication, 1989.
Gorski, T., and Gorska, K., "Inhibiting effects of full
daylight on the germination of Lactuca sativa L ."
Planta, 144 (1979): 121-124.
Hutchins, H . E., and banner, R. M., "The central role of
Clark's nutcracker in the dispersal and establishment
of whitebark pine." Oecologia, 55(1982): 192-201.
Jacobs, J., and Weaver, T., "Effects of temperature and
light levels on germination and early growth of Pinus
albicaulis under laboratory conditions." In:
Proceedings— Whitebark pine ecosystems: ecology and
management of a high-mountain resource. Ogden, UT:
U.S. Department of Agriculture, Forest Service,
Intermountain Research Station, in press, 1990.
Kahn, "The physiology and biochemistry of seed dormancy
and germination." Elsevier, North-Holland Biomedical
Press, (1977). 547 p.
Kendall, K. C., "Use of pine nuts by grizzly and black
bears in the Yellowstone area." Interagency conference
on bear research and management. In: Bears— their
biology and management. Calgary, AB: International
Association of Bear Management. 5(1983): 166-173.
Knight, R. R., Blanchard, B . M., and Mattson, D . J.,
"Yellowstone grizzly bear investigations: annual
report of the interagency study team— 1987." Bozeman,
MT: U.S. Department of the Interior, National Park
Service, 1987.
Krugman, S. L., and Jenkinson, J. L., "Pinus L . pine." In:
Schopmeyer, C. S., ed. Seeds of woody plants in the
United States. Agricultural Handbook 450. Washington,
DC: U.S. Department of Agriculture, Forest Service,
(1974): 598-638.
71
Lanner, R. M., "Avian seed dispersal as a factor in the
ecology and evolution of limber and whitebark pines."
In: Sixth North American forest biology workshop
proceedings. Edmonton, AB: University of Alberta,
(1980): 14-48.
banner, R. M., "Adaptations of whitebark pine for seed
dispersal by Clark's nutcracker." Canadian Journal of
Forest Research, 12(2) (1982): 391-402.
banner, R. M., and Vander .Wall, S . D., "Dispersal of
limber pine seed by Clark's nutcracker." Journal of
Forestry, 78(1) (1980): 637-639.
McCaughey, W. W., and Schmidt, W. C., "Autecology of
whitebark pine." In: Proceedings— Whitebark pine
ecosystems: ecology and management of a high-mountain
resource. Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountain Research Station, in
press, 1990.
Mosteller, F., and Youtz, C., "Tables of the Freeman-Tukey
transformations for the binomial and Poisson
distributions." Biometrika, 48(3 and 4)
(1961): 433-440.
Perry, David A., "Nutrients and timber harvest in pine
forests: state of knowledge." Bulletin 62. Corvallis,
OR: Oregon State University, Forestry Research
Laboratory, (1979). 37 p.
Pfister, R. D., Kovalchick, B . L., Arno, S., and Presby,
R., "Forest habitat types of Montana." General
Technical Report INT-34. Ogden, UT: U.S. Department
of Agriculture, Forest Service, Intermountain Forest
and Range Experiment Station, 1977. 174 p.
Pitel, J. A., and Wang, B . S. P., "A preliminary study of
dormancy in Pinus albicaulis seeds." Bi-Monthly
Research Notes. Chalk River, ON: Canadian Forest
Service, Petawawa National Forest Institute, 36(1)
(1980): 4-5.
SAS Institute Inc. "SAS/STAT guide for personal computers."
Version 6 Edition. Gary, NC: SAS Institute Inc.,
1987. 1028 p.
72
Schmidt, W. C., and Lotan, J. E., "Establishment and
initial development of lodgepole pine in response to
residue management. General Technical Report INT-9O.
Ogden, UT: U.S. Department of Agriculture, Forest
Service, Intermountain Forest and Range Experiment
Station, (1980): 271-286.
Schmidt, W. C., Shearer, R. C., and Roe, A. L., "Ecology
and silviculture of western larch forests." Technical
Bulletin 1520. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain Forest and
Range Experiment Station, 1976. 96 p.
Seidel, K. W., "Regeneration in mixed conifer shelterwood
cuttings in the Cascade Range of Eastern Oregon."
Research Paper PNW-264. Portland, OR: U.S. Department
of Agriculture,.Forest Service, Pacific Northwest
Forest and Range Experiment Station, 1979. 29 p.
Snedecor, G. W., arid Cochran, W. B.,. "Statistical
methods." 7th edition. Ames, IA: Iowa State
University Press, 1980. 507 p.
Soil Survey Staff, "Soil taxonomy: a basic system of soil
classification for making and interpreting soil
surveys." Agricultural Handbook No. 436. Washington,
DC: U.S. Department of Agriculture, 1975. 682 p.
Tomback, D. F., "Foraging strategies of Clark's
nutcracker." Living Bird, 16(1977): 123-160.
Tomback, D. F., "Notes on cones and vertebrate-mediated
seed dispersal." Madrono, 28(2) (1981): 91-94.
Tomback, D . F., "Dispersal of whitebark pine seeds by
Clark's nutcracker: a mutualism hypothesis." Journal
of Animal Ecology, 51 (2) (1982) : 451-467.
Tomback, D . F., "Nutcrackers and pines: coevolution or
coadaptation?" In: Niteck, M. H., ed. Coevolution.
Chicago and London: University of Chicago Press,
(1983): 179-223.
Vander Wall, S . B., and Hutchins, H . E., "Dependence of
Clark's nutcracker, Nucifraga columbiaha, on conifer
seeds during postfledgling period." Canadian FieldNaturalist, 97(2) (1983): 208-214.
Vogl, Richard J., and Ryder, Calvin, "Effects of slash
burning on conifer reproduction in Montana's Mission
Range." Northwest Science, 43(3) (1969): 135-147.
73
Weaver, T., and Dale, D., "Finns albicaulis in Central
Montana: environment, revegetation, and production."
American Midland Naturalist, 92(1974): 222-230.
Weaver. T., and Forcella, F., "Cone production in Finns
albicaulis forests." In: Shearer, R. C., compiler.
Proceedings— Conifer tree seed in the Inland Mountain
West; 1985 August 5-6; Missoula, MT. General Technical
Report INT-203. Ogden, ,UT: U.S. Department of
Agriculture, Forest Service, Intermountain Research
Station, 1986: 68-76.
Weaver, T., Forcella, F., and Dale, D., "Stand development
in whitebark pine woodlands." In: Proceedings—
Whitebark pine ecosystems: ecology and management of
a high-mountain resource. Ogden, UT: U.S. Department
of Agriculture, Forest Service, Intermountain Research
Station, in press, 1990.
Zasada, J . C., Foote, M. J., Deneke, F . J., and Parkerson,
R • H., "Case history of an excellent white spruce
cone and seed crop in Interior Alaska: cone and seed
production, germination, and seedling survival."
General Technical Report PNW-65. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific
Northwest Forest and Range Experiment Station, 1978.
53 p.
74
APPENDIX
Table 12.
Precipitation based on inches of accumulated water per month for the
hydrologic years 1987, 1988, and 1989 at the Canyon weather station in
Yellowstone National Park, Wyoming. Data supplied by the Soil
Conservation Service.
Year
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
1987 '
0.6
(33)
3.9
(155)
0.3
(12)
2.3
(78)
1.4
(65)
1.5
(64)
0.4
(21)
4.4
(167)
1.6
(61)
4.0
(222)
1.0
(47)
0.2
(9)
21.6
(78)
1988
0.4
(22)
1.9
(76)
2.6
(100)
3.3
(112)
2.5
(116)
2.5
(107)
3.3
(173)
2.4
(91)
0.5.
(19)
0.2
(U)
0.4
(19)
1.2
(51)
21.2
(76)
1989
0.2
(11)
4.1
(163)
1.7
(66)
3.8
(129)
2.0
(93)
5.7
(244)
1.6
(84)
3.2
(121)
1.3
(49)
1.4
(78)
2.0
(94)
0.7
(30)
27.7
(100)
1.80
2.51
2.59
2.95
2.16
2.34
1.91
2.64
2.63
1.80
2.13
2.34
27.81
29 yr
Average
Sept. Total
- Values in parentheses are the percent of accumulated precipitation in relation to the long-term
average for that month or total.
Table 13.
Precipitation based on inches of accumulated water per month for the
hydrologic years 1987, 1988, and 1989 at the Mill Creek weather station
on the Gallatin National Forest. Data supplied by the.Soil Conservation
Service.
Year
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept. Total
1987
0.5
(33)
1.6
(90)
0.3
(15)
1.1
(57)
0.8
(55)
2.0
(79)
1.2
(56)
4.3
(107)
1.8
(70)
3.9
(175)
3.4
(209)
0.4
(18)
21.3
1988
0.5
(33)
0.6
(34)
1.6
1.9
(99)
2.3
(158)
2.9
(114)
3.4
(157)
5.3
(132)
1.2
(47)
0.9
(40)
0.1
(<D
0.7
(32)
21.4
(78)
1.8
(120)
2.4
(135)
1.0
(49)
1.7
(89)
1.3
(89)
2.8
(HO)
1.3
(60)
6.3
(156)
2.9
(112)
2.0
(90)
1.3
(80)
1.8
(83)
26.6
1.50
1.78
2.06
1.92
1.46
2.54 . 2.16
4.03
2.58
2.23
1.63
2.18
26.06
1989
16 yr
Average
- Values in parentheses are the percent of accumulated precipitation in relation to the long-term
average for that month or total.
r
T a b l e 14
1988 analysis of variance results showing significance of biotic and
microsite factors and two-factor interactions on cumulative percent
e m e r g e n c e (arc sine of t h e s q u a r e r oot of p r o p o r t i o n transformation)
of w h i t e b a r k p i n e for e a c h r e c o r d i n g date.
.................................
Factors and
interactions
I
6/16
6/25
7/1
7/7
Predator exclusion
.1699
.0035
.0003
Shade cover
.6063
.3434
Seedbed condition
.1077
Sowing depth
Probability of 'F ' by Date
7/14
7/21
7/28
8/11
8/24
9/9
9/20
10/1
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.4613
.6064
.4770
.3384
.2901
.2752
.2521
.2483
.2483
.2483
.0102
.0139
.0223
.0190
.0091
.0147
.0199
.0221
.0055
.0055
.0055
.1077
.1323
.0005
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
Replicate
.1298
.0519
.0032
.0022
.0007
.0004
.0002
.0001
.0002
.0001
.0001
.0001
Pred x Shade
.4225
.3660
.2452
.2278
.2090
.1881
.1506
.1431
.1651
.1466
.1466
.1466
Pred x Seed
.1699
.0149
.0308
.1252
.1842
.0763
.0539
.0567
.0599
.0494
.0494
.0494
Pred x Sow
.1699
.1730
.0266
.0048
.0045
.0034
.0042
.0032
.0069
.0144
.0144
.0144
Shade x Seed
.6063
.2516
.2015
.1033
.0602
.0598
.0639
.0733
.0995
.0544
.0544
.0544
The probability values represent the significance probability associated with
the F statistic.
T a b l e 15 .
1989 analysis of v a r i a n c e results showing significance of biotic and
microsite factors and two-factor interactions on cumulative percent
e m e r g e n c e (arc s i n e o f t h e s q u a r e r o o t of p r o p o r t i o n t r a n sformation)
of w h i t e b a r k p i n e for e a c h r e c o r d i n g date.
Probability of 'F ' by Date ^
Factors and
interactions
i
6/9
6/17
6/23
7/7
Predator exclusion
.0000
.9292
.7793
Shade cover
.0000
.0425
Seedbed condition
.0000
Sowing depth
7/14
7/21
7/27
.0011
.0421
.0743
.0655
.0047
.3916
.8286
.8192
.2985
.0721
.0101
.0001
.0000
.0001
.0001
.0296
Replicate
.0000
.2090
.1176
Pred x Shade
.0000
.9027
Pred x Seed
.0000
Seed x Sow
.0000
8/4
8/18
8/31
9/11
10/19
.1883
.3825
.3828
.3110
.3110
.5298
.1537
.0679
.0849
.0080
.0080
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.0001
.1192
.1216
.0723
.0447
.0651
.0888
.0979
.0736
.0736
.9481
.7268
.5722
.3875
.2634
.1844
.1749
.1315
.1618
.1618
.6179
.7614
.6825
.7192
.5355
.2600
.1371
.0646
.0690
.0477
.0477
.2985
.0202
.0011
.0001
.0001
.0001
.0001
.0001
.0001
.0015
.0015
The probability values represent the significance probability associated with
t h e --F s t a t i s t i c .
MONTANASTATEUNIVERSITYLIBRARIES
3
IHlllIIllIfliI
762 10074746 6